Inflammatory bowel disease (IBD), Crohn's disease and ulcerative colitis, is a complex genetic disorder with both genetic and environmental causes. The distribution of disease genes varies by ethnic ancestry. Despite the large African American (AA) population with IBD in North America, determination of genetic causes in this population is only in its infancy relative to that of white and even Asian populations. We have developed the largest cohort of AA IBD cases (presently 258 confirmed cases) and ethnically matched controls. These have been enrolled into the NIDDK IBD Genetics Consortium Repository for broad use in future genetic studies. We have also analyzed the phenotype of IBD in the AA population and determined that the phenotype pattern is significantly different from that of IBD in the white population. However, we also found that IBD among AAs is frequently familial, suggesting that like the white population, there are underlying susceptibility genes. We found that unlike whites, NOD2 does not play a significant role in causing Crohn's disease in AAs. However we have replicated a significant role for the IBD5-OCTN1/2 haplotype. As predicted, linkage disequilibrium (LD) was greatly reduced for the AA population, suggesting that the IBD gene mapping in AA patients may allow more finite resolution in areas of high LD. Also, we found unique NOD2 polymorphisms suggesting that the greater genetic diversity within the AA population may provide greater knowledge of disease causing variations in the human genome. We now propose to enlarge the AA cohort to 800 cases and matched controls by year 5. Working with the consortiums DCC, we will perform a whole genome association study in Year 4, to identify IBD genes. We will also recruit an AA IBD replication population to confirm genetic findings from the GWA studies. We will use the larger population to determine the significance of unique African ancestral NOD2 variants, further reduce the IBD5 haplotype, determine the cause of IBD3 linkage and reduce its haplotype and test for association and identify unique African ancestral variants for any candidate genes identified in the white population by our NIDDK consortium collaborators. [unreadable] [unreadable]